Skill flagged — suspicious patterns detected

ClawHub Security flagged this skill as suspicious. Review the scan results before using.

LarrySkill

v1.0.0

Automate TikTok slideshow marketing for any app or product. Researches competitors, generates AI images, adds text overlays, posts via Postiz, tracks analyti...

0· 172·1 current·1 all-time
byBatsirai Chada@batsirai·duplicate of @olliewazza/larry
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Suspicious
high confidence
!
Purpose & Capability
The skill's description (TikTok slideshow marketing) matches the included scripts and references to Postiz, image generation providers, and optional RevenueCat. However, the registry metadata declares no required environment variables or config paths while the scripts expect a config.json with postiz.apiKey, imageGen.apiKey, and optional RevenueCat keys. That omission is inconsistent: a marketing/posting automation legitimately needs those credentials, so the metadata should declare them.
Instruction Scope
SKILL.md and the scripts instruct the agent to research competitors via the browser, generate images with external APIs, create and edit files (config.json, hook-performance.json, reports, analytics snapshots), write logs/webhook event files, and call third-party APIs (Postiz, OpenAI/Stability/Replicate, RevenueCat). These actions are within the stated purpose, but they involve handling and storing secrets and scraping/dashboard access — the instructions also direct the agent to 'IMMEDIATELY start a conversation' on first load, which may be surprising behavior to users.
Install Mechanism
No install spec (instruction-only with shipped scripts) — lowest install-risk for package downloads. However, node-canvas is a native module that may require build tools (Python, make, C++ compiler), and the SKILL.md explicitly tells the agent to install Node and native dependencies; users should be prepared for that build complexity.
!
Credentials
The scripts require sensitive credentials (Postiz API key in config.postiz.apiKey, image generation API keys for OpenAI/Stability/Replicate, optional RevenueCat secret keys) but the skill manifest declares no required env vars or config paths. That mismatch is a red flag: the skill will need secrets to operate, and the registry metadata currently fails to surface that requirement to the user.
Persistence & Privilege
always is false and the skill does not request system-wide configuration changes. It does instruct an immediate onboarding conversation on load and expects to run recurring cron-like jobs (daily-report, analytics checks) under user control. Autonomous invocation is allowed by default (not flagged here), so consider whether you want the agent to run periodic tasks and API calls without manual confirmation.
What to consider before installing
What to check before installing: - Expect to supply API keys and a config file: the scripts require a Postiz API key and an image-generation API key (OpenAI/Stability/Replicate) and optionally RevenueCat secrets; the registry metadata currently does NOT declare these — confirm where and how you'll store these credentials (config.json vs env vars) and ensure the skill's manifest is updated to reflect required secrets. - Review where secrets are written/read: scripts read/wrote config.json and local JSON logs (analytics-snapshot.json, hook-performance.json, rc-events.json). Decide whether storing secrets and webhook logs on disk is acceptable for your environment. - Validate third-party services: verify Postiz (postiz.pro) and any recommended image provider endpoints and terms before handing over API keys. Consider using limited-scope keys or test accounts. - Be prepared for native builds: node-canvas is a native module that may need build tools; test in an isolated environment or container first. - Control autonomous actions: the skill asks to start a conversation immediately and can run scheduled scripts; if you don't want automatic posting/analytics or autonomous agent invocation, restrict the agent or run scripts manually until you audit behavior. - If unsure, run manually: execute the included Node scripts locally with test credentials in a sandbox to confirm behavior before enabling the agent to operate them automatically.
!
scripts/check-analytics.js:45
File read combined with network send (possible exfiltration).
!
scripts/daily-report.js:40
File read combined with network send (possible exfiltration).
!
scripts/generate-slides.js:38
File read combined with network send (possible exfiltration).
!
scripts/post-to-tiktok.js:30
File read combined with network send (possible exfiltration).
About static analysis
These patterns were detected by automated regex scanning. They may be normal for skills that integrate with external APIs. Check the VirusTotal and OpenClaw results above for context-aware analysis.

Like a lobster shell, security has layers — review code before you run it.

latestvk978kj1rtmrhdxqh7wsrn90a8d82vqg9

License

MIT-0
Free to use, modify, and redistribute. No attribution required.

Comments